SOAP: Efficient Feature Selection of Numeric Attributes

The attribute selection techniques for supervised learning, used in the preprocessing phase to emphasize the most relevant attributes, allow making models of classification simpler and easy to understand. Depending on the method to apply: starting point, search organization, evaluation strategy, and...

Descripción completa

Detalles Bibliográficos
Autores: Ruiz Sánchez, Roberto, Aguilar Ruiz, Jesús Salvador, Riquelme Santos, José Cristóbal
Tipo de recurso: capítulo de libro
Estado:Versión publicada
Fecha de publicación:2002
País:España
Institución:Universidad de Sevilla (US)
Repositorio:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/39157
Acceso en línea:http://hdl.handle.net/11441/39157
https://doi.org/10.1007/3-540-36131-6_24
Access Level:acceso abierto
Palabra clave:Artificial Intelligence (incl. Robotics)
Computation by Abstract Devices
id ES_f8260ed6688ed8d2d4b3b5bc5cc46fbb
oai_identifier_str oai:idus.us.es:11441/39157
network_acronym_str ES
network_name_str España
repository_id_str
spelling SOAP: Efficient Feature Selection of Numeric AttributesRuiz Sánchez, RobertoAguilar Ruiz, Jesús SalvadorRiquelme Santos, José CristóbalArtificial Intelligence (incl. Robotics)Computation by Abstract DevicesThe attribute selection techniques for supervised learning, used in the preprocessing phase to emphasize the most relevant attributes, allow making models of classification simpler and easy to understand. Depending on the method to apply: starting point, search organization, evaluation strategy, and the stopping criterion, there is an added cost to the classification algorithm that we are going to use, that normally will be compensated, in greater or smaller extent, by the attribute reduction in the classification model. The algorithm (SOAP: Selection of Attributes by Projection) has some interesting characteristics: lower computational cost (O(mn log n) m attributes and n examples in the data set) with respect to other typical algorithms due to the absence of distance and statistical calculations; with no need for transformation. The performance of SOAP is analysed in two ways: percentage of reduction and classification. SOAP has been compared to CFS [6] and ReliefF [11]. The results are generated by C4.5 and 1NN before and after the application of the algorithms.Lenguajes y Sistemas Informáticos2002info:eu-repo/semantics/bookPartinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/11441/39157https://doi.org/10.1007/3-540-36131-6_24reponame:idUS. Depósito de Investigación de la Universidad de Sevillainstname:Universidad de Sevilla (US)InglésAdvances in Artificial Intelligence — IBERAMIA 2002, Lecture Notes in Computer Science, Volume 2527, pp 233-242 (2002)info:eu-repo/semantics/openAccessoai:idus.us.es:11441/391572026-06-17T12:51:07Z
dc.title.none.fl_str_mv SOAP: Efficient Feature Selection of Numeric Attributes
title SOAP: Efficient Feature Selection of Numeric Attributes
spellingShingle SOAP: Efficient Feature Selection of Numeric Attributes
Ruiz Sánchez, Roberto
Artificial Intelligence (incl. Robotics)
Computation by Abstract Devices
title_short SOAP: Efficient Feature Selection of Numeric Attributes
title_full SOAP: Efficient Feature Selection of Numeric Attributes
title_fullStr SOAP: Efficient Feature Selection of Numeric Attributes
title_full_unstemmed SOAP: Efficient Feature Selection of Numeric Attributes
title_sort SOAP: Efficient Feature Selection of Numeric Attributes
dc.creator.none.fl_str_mv Ruiz Sánchez, Roberto
Aguilar Ruiz, Jesús Salvador
Riquelme Santos, José Cristóbal
author Ruiz Sánchez, Roberto
author_facet Ruiz Sánchez, Roberto
Aguilar Ruiz, Jesús Salvador
Riquelme Santos, José Cristóbal
author_role author
author2 Aguilar Ruiz, Jesús Salvador
Riquelme Santos, José Cristóbal
author2_role author
author
dc.contributor.none.fl_str_mv Lenguajes y Sistemas Informáticos
dc.subject.none.fl_str_mv Artificial Intelligence (incl. Robotics)
Computation by Abstract Devices
topic Artificial Intelligence (incl. Robotics)
Computation by Abstract Devices
description The attribute selection techniques for supervised learning, used in the preprocessing phase to emphasize the most relevant attributes, allow making models of classification simpler and easy to understand. Depending on the method to apply: starting point, search organization, evaluation strategy, and the stopping criterion, there is an added cost to the classification algorithm that we are going to use, that normally will be compensated, in greater or smaller extent, by the attribute reduction in the classification model. The algorithm (SOAP: Selection of Attributes by Projection) has some interesting characteristics: lower computational cost (O(mn log n) m attributes and n examples in the data set) with respect to other typical algorithms due to the absence of distance and statistical calculations; with no need for transformation. The performance of SOAP is analysed in two ways: percentage of reduction and classification. SOAP has been compared to CFS [6] and ReliefF [11]. The results are generated by C4.5 and 1NN before and after the application of the algorithms.
publishDate 2002
dc.date.none.fl_str_mv 2002
dc.type.none.fl_str_mv info:eu-repo/semantics/bookPart
info:eu-repo/semantics/publishedVersion
format bookPart
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11441/39157
https://doi.org/10.1007/3-540-36131-6_24
url http://hdl.handle.net/11441/39157
https://doi.org/10.1007/3-540-36131-6_24
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Advances in Artificial Intelligence — IBERAMIA 2002, Lecture Notes in Computer Science, Volume 2527, pp 233-242 (2002)
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.source.none.fl_str_mv reponame:idUS. Depósito de Investigación de la Universidad de Sevilla
instname:Universidad de Sevilla (US)
instname_str Universidad de Sevilla (US)
reponame_str idUS. Depósito de Investigación de la Universidad de Sevilla
collection idUS. Depósito de Investigación de la Universidad de Sevilla
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869424975212969984
score 15.300719

Similares en LA Referencia